DESCRIPTION: (Applicant's Abstract) Mechanisms of drug addiction appear to be remarkably conserved. Work in animal model systems can therefore greatly inform our understanding of human addiction. The economy-of-scale and molecular-genetic tools available in fruit flies will be used to identify genes and genetic pathways of addiction. Recent studies in Drosophila have established behavioral sensitivity models for both cocaine and ethanol. In the latter, an allele of amnesiac, a learning and memory gene that has some homology to a vertebrate neuropeptide that stimulates cAMP signaling, was identified. Amnesiac mutants show decreased tolerance to ethanol, a defect that can be reversed with pharmacological stimulation of cAMP levels. In the former, several single-gene mutants were shown to affect cocaine sensitization, which also was mediated by dopaminergic signaling. These studies validate Drosophila as a genetic model system for both cocaine and alcohol addiction and implicate the cAMP signaling pathway in the response to both. This Drosophila model system of addiction will be enhanced with the use of DNA microarray chips to identify drug-related changes in gene expression on a genomic-wide scale. In preliminary experiments, a partial genome prototype Drosophila chip has been used to identify 176 candidate genes transcriptionally altered in amnesiac mutants and 75 that are transcriptionally altered by pharmacological manipulation of cAMP levels. We have confirmed some of the statistical candidates from the chip using either Northern blot analysis or quantitative PCR. From initial analyses of more than 100 chips, we also have developed a novel and general statistical method to analyze DNA chip data. These initial results now will be integrated with data from exposures to cocaine and ethanol. Central to our approach is the comparison of three genomic responses: (1) drug-exposed vs. control, (2) pharmacological manipulation vs. control and (3) mutant vs. wild-type. Chosen first for further in vivo study will be those genes that are transcriptionally responsive in all three contexts - and to both cocaine and ethanol exposures. Given the "homology of function" already observed for drug responses in flies and vertebrates, the genes identified in flies most likely will have homologues in vertebrates. Hence this work will lead directly to work in mammalian models of addiction and ultimately will yield a greater understanding of the genetic basis of addiction in humans.